MOBILE & WIRELESS HEALTH

Cloud Computing Mobile Application for Remote Monitoringof Bell’s Palsy

P Watts1 & P Breedon1& C Nduka2,3 & C Neville2 & V Venables2 & S Clarke3

Received: 17 April 2019 /Accepted: 15 July 2020# The Author(s) 2020

AbstractMobile applications provide the healthcare industry with a means of connecting with patients in their own home utilizing theirown personal mobile devices such as tablets and phones. This allows therapists to monitor the progress of people under their carefrom a remote location and all with the added benefit that patients are familiar with their own mobile devices; thereby reducingthe time required to train patients with the new technology. There is also the added benefit to the health service that there is noadditional cost required to purchase devices for use. The Facial Remote Activity Monitoring Eyewear (FRAME) mobileapplication and web service framework has been designed to work on the IOS and android platforms, the two most commonlyused today. Results: The system utilizes secure cloud based data storage to collect, analyse and store data, this allows for near realtime, secure access remotely by therapists to monitor their patients and intervene when required. The underlying framework hasbeen designed to be secure, anonymous and flexible to ensure compliance with the data protection act and the latest General DataProtection Regulation (GDPR); this new standard came into effect in April 2018 and replaces the Data Protection Act in the UKand Europe.

Keywords Cloud computing .Mobile application . Bell’s palsy . Healthcare solution .Wearable technology . System design

Introduction

Background

Bell’s palsy represents 60% of all facial palsy cases current inthe UK, with up to 25,640 new cases in the UK annually [1].The annual incidence rate within the UK population is esti-mated to be 20 per 100,000 and an estimated 1 in 60 peopleare often left with a permanent disability with an estimated100,000 people living in the UKwith permanent facial palsy’s[2]. Bell’s palsy has a significant impact on patients’ lives

affecting their facial function, appearance and communica-tion. This often affects a patient’s ability to speak, eat withoutdribbling or blink and is shown to have a significant physio-logical impact on patients leading to conditions such as de-pression and anxiety [3, 4].

Facial muscles are different to other skeletal muscles in thatthey have poor proprioceptive feedback. Most people are un-able to control individual facial muscles, such as raising alter-nating eyebrows. This is by design; the face’s role is that of amethod to transact trusted interactions, which is aided by lim-ited conscious control of expressions.

Following injury to the facial nerve, the degree of recoverydepends on the severity of the injury which is typically clas-sified by a grading system such as the House Blackman Scale,or the Sunnybrook Facial Grading System [5]. Studies showthat early intervention using specialized exercises has a sig-nificant impact on improving the patient’s condition [6]. Thelack of proprioception makes recovery particularly difficult aspatients are usually unaware of abnormal movements theirfaces exhibit during daily activities. Without additional feed-back their facial function may worsen and potentially developinto permanent abnormal movements.

This article is part of the Topical Collection on Mobile and WirelessHealth

* P Breedonphilip.breedon@ntu.ac.uk

1 Medical Engineering Design Research Group, Nottingham TrentUniversity, Nottingham, UK

2 Queen Victoria Hospital, NHS Trust, East Grinstead, West Sussex,UK

3 Emteq Ltd, Sussex Innovation Centre, Brighton, UK

https://doi.org/10.1007/s10916-020-01605-7

/ Published online: 29 July 2020

Journal of Medical Systems (2020) 44: 149

Biofeedback for facial paralysis is well established in spe-cialist centres [7, 8, 9, 10], however, intermittent biofeedbackis of limited value. Mobile technology providing real-timefeedback and guidance to patients and therapists promises tosignificantly improve in the rehabilitation of facial paralysis[11].

Related works

The research sector contains many existing applications anddesigns for remote home healthcare and throughout the pro-ject the team considered the research in its approach to thedesign and implementation of the FRAME project.

The CARA (Context-Aware Real-time Assistant) projectled by Bingchuan Yuan at the University College Cork [12]researched the utilization of wireless remote monitoring ofpatients via a web browser to allow a healthcare professionalor carer to monitor their patients remotely and from any webenabled device.

A team at Edinburgh university led by E Ekonomou in2011 presented a project on an Integrated Cloud-basedHealthcare Infrastructure [13] (DACAR) [X2], this projectprioritized high levels of security and privacy whilst utilizinga secure cloud based environment. This is a key priority in thedevelopment of any cloud based solution like the FRAMEproject.

There are a number of related research projects looking intothe use of Electromyography (EMG) to detect the movementof facial muscles, a paper by Jun-Wen Tan from theUniversity of Ulm in 2012 investigated the use of EMG todetect the activity over two key muscles in the face to identifyvarious emotional responses [14]. Research by A Van Boxtelat Tilburg University [15] investigated the location of EMGsensors and the methodological aspects of recording the facialEMG signals to identify movements in the key facial muscles.

Overview

The FRAME project required the development of an onlinecloud based mobile solution for the treatment and monitoringof patients suffering from Bell’s palsy, the project consists ofthree key parts all of which make up a framework that comestogether to form the final end user product (Fig. 1).

The Facial Remote Activity Monitoring Eyewear(F.R.A.M.E) mobile application is an application designed towork on IOS and Android for the remote care of individualsduring their treatment. The application contains information andexercises to aid in the patient’s rehabilitation from Bell’s palsy.

An online web application provides a cloud basedRepresentational State Transfer (REST) application-program-ming interface (API) hosted on the Microsoft Azure cloudplatform for secure transmission of data across multipledevices.

A therapist console built into the web application providesaccess for permitted therapists (via a web browser) to accesspatient data remotely. This allows for the remote monitoringof patients progress over a period of weeks, months or years.

The mobile interfaces with two wearable hardware devicesdeveloped as part of the FRAME project via Bluetooth LowEnergy (BLE) to provide sensor feedback from the patientwhilst they are performing exercises as instructed by the mo-bile application.

Design requirements

Design of the system is split into three key areas to ensure thatthe developed system addresses the needs of the patients andtherapist whilst also ensuring the system complied with secu-rity requirements and GDPR regulations. The first area (Table1) addresses the patient requirements (PR) of the system, theserequirements focus on the functionality within the patient fac-ing mobile application.

The second area (Table 2) identified addresses theTherapist Requirements (TR) and focuses on the web consoleand the ability for therapists to monitor their patients through-out treatment (TR2, TR6). Therapists highlighted the need forinformation about Bell’s palsy to be readily available to pa-tients within the mobile application (TR4, TR5).

The third area (Table 3) addresses the SystemRequirements (SR) to ensure data is secured and access is onlygranted to authorised personnel (SR1, SR2). It also ensuresthat the designed system complies with data protection andGDPR regulations (SR3).

System framework

The F.R.A.M.E application framework consists of a group ofsystem modules running on multiple platforms, all connectedto one another to form the core of the F.R.A.M.E systemframework. Each individual system module works closelywith the other system modules to create a comprehensive,flexible and secure solution.

Mobile application

The mobile application is designed and developed to work onIOS and Android (PR1, TR1), allowing deployment onto themajority of mobile devices currently on the market; alsoallowing deployment onto personal devices of the majorityof patients receiving treatment (TR1). The mobile applicationutilizes the Xamarin Cross Platform programming language toallow for simultaneous development of the application forboth platforms, sharing common code between the two plat-forms. This shared code speeds up development by reducing

149 Page 2 of 9 J Med Syst (2020) 44: 149

the time to develop separate solutions and thereby reducescosts of overall development.

The aim of the application is to assist patients by providingessential information and advice helping them manage theircondition daily. Information has been taken from current NHSguidance handouts and DVD video resources (TR5) and hasbeen integrated into categorized sections of the application(PR3, TR4). The application provides animated guidanceand timings for performing daily stretch and massage exer-cises, aiding patient’s recovery and rehabilitation. Massageand stretch guidance is provided by a set sequence of exerciseswith accompanying animations all timed to be copied in linewith current NHS guidelines (PR4, TR4). An example ofscreens from the app can be seen in Fig. 2. The patient canperform these exercises at any time of day (PR6) with themobile application sending usage statistics to the cloud serverfor observation and monitoring by the therapist (TR2).

The mobile application connects to the F.R.A.M.E servervia any available internet connection on the device, allowingfor the retrieval and transmission of data between the serverand client. Data is stored on the cloud server and not thedevice; this allows a patient to swap seamlessly between mul-tiple devices if required (PR2). This is an important designaspect when developing for applications designed for use overextended periods, users can own multiple devices and devicescan fail and/or be replaced during treatment. For security, the

application requires entry of a secure username and passwordchosen by the user during registration.

F.R.A.M.E EMG goggles

The F.R.A.M.E project involved the development a hardwaredevice to assist in the rehabilitation of patients and to providefeedback to the therapist throughout treatment [9, 10]. Thedevice connects to the mobile application via Bluetooth LowEnergy (BLE).

The hardware takes the format of a pair of goggles withintegrated Electromyography (EMG) sensors places at 10 keylocations around the patients face to allow the collection andmonitoring of key facial muscles responsible for expressions(Fig.3) [9, 10]. EMG data is streamed to the mobile applica-tion via BLE where it is stored on the device prior to beinguploaded to the server using a RESTful API.

The application currently consists of an exercise routinecontaining two exercises chosen to best test the mask and itsusability during real world conditions. Patients facial musclesneed to be relaxed before performing any voluntary facialexpression to prevent strain and further damage, with this inmind the first exercise is designed to help the user relax theirfacial muscles (PR7) and to make them aware of tension inany of their facial muscles (Fig. 4, right). The second exerciseinstructs the user to perform a gentle balanced smile (PR8)

Fig. 1 The FRAME system overview

Table 1 Patient requirements (PR)

PR1 Mobile application for use on personal mobile devices (Apple/Android)

PR2 Ability to use multiple devices throughout treatment such as multiple phones or tablets

PR3 To provide information and advice on living with Bell’s palsy.

PR4 Provide guidance on daily exercises and stretches

PR5 Provide mirror therapy within the mobile application

PR6 Ability to perform exercises daily when it suits them so as to fit in with other daily activities

PR7 Visual guidance with feedback for muscle relaxation whilst using the EMG mask

PR8 Visual guidance with feedback whilst performing a gentle balanced smile with the EMG mask

Page 3 of 9 149J Med Syst (2020) 44: 149

whilst focusing on achieving an even and balance smile be-tween the left and right side of their facial muscles (Fig. 4,left). Should the user exhibit an unbalanced smile the EMGsensors will pick up on the variation and provide user feed-back on the mobile application with on screen instructions tothe user to relax and gently try again. The smile exercise waschosen by therapists on the team to best validate the maskwitha vision to expand the number of exercises once validation hasbeen completed.

Cloud web application

The web application is deployed on theMicrosoft Azure cloudplatform and is written in C# using the MVC 5 framework.The purpose of the web application is to provide the imple-mentation of the RESTful API for data transmission and toprovide the therapist and administrator web portals accessiblevia any web browser. The web application utilises a secureAzure hosted MSSQL database for storage of linked data setsand blob storage for the storage of larger chunks of data suchas files and images.

RESTful API

The RESTful [16] web API has been developed to communi-cate with the mobile application using JavaScript ObjectNotation (JSON) [17]. This is a lightweight machine-readabledata interchange format. Data is passed between the mobileapplication and web application via a Secure Sockets Layer(SSL) [18] connection utilising Hypertext Transfer Protocol(HTTP) [19] POST commands. All content is encrypted aspart of the SSL layer in order to prevent the data beingintercepted, read or modified during transmission.

Therapist portal

A therapist portal is built into the web application as a websiteand is accessed from any web browser on a mobile device,Windows computer or Mac computer (TR3). The websiteprovides a secure interface for registered therapists to viewthe progress of patients under their care. Access to the websiterequires therapists to login using a valid username and pass-word (SR2). The main console allows the therapist to see aquick overview of patients under their care and they can seerecent activity for each patient with inactive patients clearlyindicated (TR2). The purpose of this view is to highlight pa-tients who are not currently active in their treatment allowingthe therapist to address key concerns and intervene shouldthey see fit. The therapist has the option to see more detailsabout each patient via a ‘Patient Console’ link. The patientconsole is designed to show information relating to an indi-vidual patient and shows the therapist the patient’s currentactivity level and data collected from exercises performed bythe patient during treatment (TR2). The therapist can accessand analyze the data when needed (TR6). A visual represen-tation of the EMG data collected by the mobile applicationfrom mask device is drawn as a graph showing the muscleactivity of the patient throughout the exercise. Figure 5 showsan example of the patient console within the therapist portal.

The console also provides access to this data in the form ofdownload links for files in raw or Comma-separated values(CSV). CSV files can be opened in many common softwarepackages such as Excel and allows for further analysis by thetherapist should it be needed (TR6). The contents of these filesis anonymized for security and data protection purposes.

Security

Account login

In order for a patient to access the system, they are required tologin via the mobile application using a username and pass-word. The login system has been developed using a token-based authentication system to reduce the possibility of databeing stolen and usernames and passwords being compro-mised (SR1).

Table 2 Therapist requirements (TR)

TR1 Low cost solution for the NHS

TR2 Ability to see patient’s activity and adherence during treatment

TR3 Access to patient data from multiple devices/computers within the NHS

TR4 A digital and easy to find resource to provide NHS information/advice on living with Bell’s palsy.

TR5 Provide easy access to existing video content for managing Bell’s palsy

TR6 Provide access to data from EMG mask exercises

Table 3 System requirements (SR)

SR1 Secure login with encryption of username and passwords

SR2 Secure access to data for authorized personnel only

SR3 Anonymous data storage

SR4 24/7 Access to data and services

149 Page 4 of 9 J Med Syst (2020) 44: 149

Token based authentication works by registering a deviceto the user account on the server system and issuing a uniqueand randomized token. The first time the application is startedthe user is asked to enter their username and password, this isencrypted using AesCbcPkcs7 encryption before being trans-mitted to the server via the API, utilizing a secure SSLencrypted Post command. The username and password are

validated on the server and a Globally Unique Identifier(GUID) is assigned as a unique token to the user accountand device id. This token is sent back to the mobile applicationand the application stores the token, at no point is the pass-word stored on the mobile device. All transactions made to theserver after authentication will utilize the token along with thedevice id. This removes the need to pass the username andpassword between the device and server system during eachAPI request and removes the need to store the password on thedevices itself. The full process is shown in Fig. 6 below.

Data access

For security reasons the web portal also uses the token authen-tication to validate the therapist is logged in before streamingany images or files to the end users browser (SR2). Files arenot directly available and cannot be downloaded via a stan-dard URL request; the web browser requests the image or filethrough a secure POST command, the user’s login is validatedand the web server accesses the file securely beforeforwarding data to the browser through a secure SSLencrypted connection.

Image access and storage

To facilitate the secure storage of images on the server, en-cryption was added to all images uploaded from the applica-tion. Images are encrypted using Rijndael encryption [20] uti-lizing randomized keys known only to the server. Encryptionkeys are stored in a secure table not accessible outside of theservers secure network and are stored using a randomized IDto obfuscate the image within the database. Encrypted imagedata is then stored in secure cloud storage located inside the

Fig. 2 Screenshots from the F.R.A.M.E App. From left to right, patient information, exercise instructions, animated stretch and animated massageexercises

Fig. 3 Right side of the EMG mask sensor placements withcorresponding movement indicators for each sensor

Page 5 of 9 149J Med Syst (2020) 44: 149

isolated private network accessible only by the web applica-tion. Additionally all data stored within Azure Cloud Storageis itself encrypted by the Azure service using 256 bit AESencryption [21].

To access the image a request needs to be made using avalid token and the image id. Validation is made on the serverto ensure the user has permission to access the image beforethe image is decoded using the secure database keys andreturned as a decoded image for use in the application orweb browser view. (Fig. 7).

GDPR and anonymizing patient data

Throughout development of the project, the team carefully con-sidered security and compliance with data privacy and the an-onymity of registered patients. In April 2018, Europe intro-duced the new General Data Protection Regulation (GDPR)rules to replace the data protection act. These rules protectmembers of the public from having personal data used withouttheir explicit consent and restricts the capability of companiesto store personal data without consent from the user.

Fig. 5 Example of patientconsole as seen by the logged intherapist

Fig. 4 EMG relaxation and gentle smile exercise within the mobile application

149 Page 6 of 9 J Med Syst (2020) 44: 149

It was identified during development that the system couldfunction without the need to store personal details that wouldallow the identification of an individual on the system. Thesystem has no need to know the identity, age or gender of anyperson using the application. This has no impact on the out-come of the treatment. Using a registration system requiring

only a username and password allows user data to be collect-ed, stored and processed anonymously without knowing theidentity of the account holder.

Therapists can track a user’s progress through the anony-mous usernames displayed on the therapist console. If desired,the therapist can seek verbal permission from an individual to

Fig. 7 Image Request Data Decryption Flow

Fig. 6 Token Authentication Security Flow

Page 7 of 9 149J Med Syst (2020) 44: 149

disclose their username, which would provide consent for theindividual patient to be identified by the therapist. This infor-mation is beyond the scope of the server system as no identi-fiable data is stored and will be the responsibility of the ther-apist to ensure GDPR compliance is adhered too.

Conclusion

Personal mobile devices and cross platform application devel-opment tools allow for the development of low cost medicalintervention applications to aid and assist a wide range ofpatients suffering long term medical conditions. Office forNational Statistics mobile phone data 2018 to 2017 reportsindicate that there are over 83 million mobile phone subscrip-tions with 93% of adults owning such a device [22]. Thehealthcare sector would benefit greatly by tapping into thisresource to both cut costs and use these platforms to provideexpert advice to members of the public through official andendorsed mobile applications.

The F.R.A.M.E application has been designed and devel-oped to include the latest advice from healthcare specialistsworking within the NHS. This help and advice would normal-ly be distributed using paper based documents within the clin-ic and these handouts can often be misplaced or forgotten bythe patient once they leave the clinic. The F.R.A.M.E mobileapplication includes this information and is a single point ofreference whilst also providing visual guidance on exercisesand stretches. These exercises guide correct compliance andadherence to the exercise routines to improve clinical out-comes and benefits of the treatment. Information can be up-dated regularly to provide the latest advice and informationfrom experts within the NHS through application updates andupdates to the information stored on the cloud platform.

Utilization of cloud based web applications allows for re-mote monitoring of patients and is available 24 h a day, 7 daysa week, and 365 days a year. Allowing patients to performtheir exercises when it suites them and providing informationand advice as and when they need it. Patients adhering to theseexercise routines and advice will receive the best treatmentavailable whilst being able to carry on with their everydaylives uninterrupted.

A therapist console has been integrated into the solution toallow therapists to regulate their workload and efficientlymanage their time, focusing on patients with the greatest needas identified through the therapist online portal. Therapists canview the activity of their patients at any time, withoutinterrupting the patient’s everyday activities, allowing for apassive treatment method with intervention only supplied asand when needed.

When designing application frameworks for use in thehealthcare sector careful consideration is needed to ensure

GDPR compliance. GDPR data protection laws require thatit is necessary to collect only user details that is essential to thetreatment process. F.R.A.M.E demonstrates that it is feasiblefor an online treatment system to be designed, developed anddeployed with minimal personally identifiable data beingrequired.

The team envisions further developing the solution toinclude more exercises into the routine and to increasethe monitoring functionality of the cloud solution to en-sure that the best treatment is achieved. Further refine-ment in the design of the hardware will improve thesensitivity of the mask and allow for compatibility withdifferent facial morphologies. A second hardware deviceis currently in development to aid patients at the nextstage of treatment. Patients suffering with Synkinesiswill be able to wear a pair of electronic glasses whichalerts them to their condition to allow for voluntarychanges in their behaviour.

Acknowledgements This project is funded by the NIHR Invention forInnovation Programme (i4i). The views expressed are those of the authorsand not necessarily those of the NHS, the NIHR or the Department ofHealth, UK.

Funding This project is funded by the NIHR Invention for InnovationProgramme (i4i). Grant number II-LA-0814-20008.

Compliance with ethical standards

Conflict of interest Paul Watts declares that he has no conflicts ofinterest.

Philip Breedon declares that he has no conflicts of interest.Charles Nduka declares that he has no conflicts of interest.Catriona Neville declares that she has no conflicts of interest.Vanessa Venables declares that she has no conflicts of interest.Simon Clarke declares that he has no conflicts of interest.

Ethical approval This article does not contain any studies with animalsperformed by any of the authors.

Informed consent was obtained from all individual participants in-cluded during studies conducted as part of this project.

All procedures performed in studies involving human participantswere in accordance with the ethical standards of the institutional and/ornational research committee and with the 1964 Helsinki declaration andits later amendments or comparable ethical standards. In the case of hu-man participation.

Open Access This article is licensed under a Creative CommonsAttribution 4.0 International License, which permits use, sharing, adap-tation, distribution and reproduction in any medium or format, as long asyou give appropriate credit to the original author(s) and the source, pro-vide a link to the Creative Commons licence, and indicate if changes weremade. The images or other third party material in this article are includedin the article's Creative Commons licence, unless indicated otherwise in acredit line to the material. If material is not included in the article'sCreative Commons licence and your intended use is not permitted bystatutory regulation or exceeds the permitted use, you will need to obtainpermission directly from the copyright holder. To view a copy of thislicence, visit http://creativecommons.org/licenses/by/4.0/.

149 Page 8 of 9 J Med Syst (2020) 44: 149

References

1. Holland, N. J. and Weiner, G. M., Recent developments in Bell’spalsy. Bmj. 329(7465):553-557, 2004.

2. Julian, D., and Partridge, J., The Incidence and Prevalence ofDisfigurement: Changing faces, London, (2008).

3. Morales, D. R., Donnan, P. T., Daly, F., Van Staa, T. and Sullivan,F. M., Impact of clinical trial findings on Bell’s palsy managementin general practice in the UK 2001–2012: interrupted time seriesregression analysis. BMJ Open. 3(7):p.e003121, 2013.

4. Weir, A. M., Pentland, B., Crosswaite, A., Murray, J. andMountain, R., Bell’s palsy: the effect on self-image, mood stateand social activity. Clin. Rehabil. 9(2):121-125, 1995.

5. Fattah, A. Y., Gurusinghe, A. D., Gavilan, J., Hadlock, T. A.,Marcus, J. R., Marres, H., Nduka, C. C., Slattery, W. H. andSnyder-Warwick, A. K., Facial nerve grading instruments: system-atic review of the literature and suggestion for uniformity. Plast.Reconstr. Surg. 135(2):569-579, 2015.

6. Manikandan, N., Effect of facial neuromuscular re-education onfacial symmetry in patients with Bell’s palsy: a randomized con-trolled trial. Clin. Rehabil. 21(4):338-343, 2007.

7. Dalla Toffola, E., Bossi, D., Buonocore, M., Montomoli, C.,Petrucci, L. and Alfonsi, E., Usefulness of BFB/EMG in facialpalsy rehabilitation. Disabil. Rehabil. 27(14):809-815, 2005.

8. Ross, B., Nedzelski, J. M. and McLean, J. A., Efficacy of feedbacktraining in long-standing facial nerve paresis. Laryngoscope 101(7):744-750, 1991.

9. Breedon, P., Watts, P., Cox, G., Szczepura, A., Baines, D.,Johnson, K., Venables, V. and Nduka, C., F.R.A.M.E – FacialRemote Activity Monitoring Eyewear. Proceedings of the IADISInternational Conference: Interfaces and Human ComputerInteraction 2017, Lisbon, Portugal, 21-23 July 2017. pp. 207-211.ISBN 9789898533647, 2017.

10. Breedon, P., Watts, P., Clarke, S., Cox, G., and Nduka, C.,F.R.A.M.E. (Facial Remote Activity Monitoring Eyewear) - Real-time patient feedback through facial expression utilising wearabletechnology, Proc. 12th Intl Conf. on Disability, Virtual Reality andAssoc. Technologies, PJ Standen, SVC Cobb, DJ Brown, PGamito, K Appiah (Eds), pp. n–m, Nottingham, England, 4-6Sept. 2018, 2018.

11. Nicastri, M. et al, Efficacy of early physical therapy in severe Bell’spalsy: |A randomized controlled trial.Neurorehabil. Neural Repair.27(6), 542–51, 2013. https://doi.org/10.1177/1545968313481280.

12. Yuan, B. and Herbert, J., Web-based real-time remote monitoringfor pervasive healthcare. In 2011 IEEE International Conference onPervasive Computing and CommunicationsWorkshops (PERCOMWorkshops) (pp. 625-629). IEEE, 2011, March.

13. Ekonomou, E., Fan, L., Buchanan, W. and Thuemmler, C., Anintegrated cloud-based healthcare infrastructure. In 2011 IEEEThird International Conference on Cloud Computing Technologyand Science (pp. 532-536). IEEE, 2011, November.

14. Tan, J. W., Walter, S., Scheck, A., Hrabal, D., Hoffmann, H.,Kessler, H. and Traue, H. C., Repeatability of facial electromyog-raphy (EMG) activity over corrugator supercilii and zygomaticusmajor on differentiating various emotions. J. Ambient. Intell.Humaniz. Comput., 3(1), pp.3-10, 2012.

15. Van Boxtel, A., Facial EMG as a tool for inferring affective states.In Proceedings of measuring behavior (pp. 104-108). Wageningen:Noldus Information Technology, 2010, August.

16. Pautasso, C., RESTful web services: principles, patterns, emergingtechnologies. In Web Services Foundations (pp. 31-51). Springer,NY: New York, 2014.

17. Ecma International, The JSON Data Interchange Syntax http://www.ecma-international.org/publications/files/ECMA-ST/ECMA-404.pdf. , 2017.

18. Freier, A., Karlton, P., Kocher, P., The secure sockets layer (SSL)protocol version 3.0 No. RFC 6101, 2011.

19. Fielding, R., Gettys, J., Mogul, J., Frystyk, H., Masinter, L., Leach,P., Berners-Lee, T., Hypertext transfer protocol–HTTP/1.1 (No.RFC 2616), 1999.

20. Daemen, J., Rijmen, V., The design of Rijndael: AES-the advancedencryption standard. Berlin, Heidelberg, Springer Science &Business Media, 2013.

21. Kaufman, C., Venkatapathy, R., Windows Azure™ SecurityOverview. http://download.microsoft.com/download/0/0/B/00BB2908-0FC2-4CDE-B370-B9D70C7A84FC/03f_WindowsAzureSecurity_Overview.pdf. , 2010.

22. Office for National Statistics (ONS), Using mobile phone data withincommuting flows. https://www.ons.gov.uk/file?uri=/aboutus/transparencyandgovernance/freedomofinformationfoi/mobilephonedata2016to2017/document1mobilephonedatareport.pdf ,2016.

Publisher’s Note Springer Nature remains neutral with regard to jurisdic-tional claims in published maps and institutional affiliations.

Page 9 of 9 149J Med Syst (2020) 44: 149